[Intel-gfx] [PATCH v2 5/6] drm/i915: Update connector_mask during readout.

[Daniel Vetter]

On Tue, Jan 05, 2016 at 10:05:21AM +0100, Maarten Lankhorst wrote:
> Op 05-01-16 om 09:35 schreef Daniel Vetter:
> > On Mon, Jan 04, 2016 at 12:53:19PM +0100, Maarten Lankhorst wrote:
> >> Signed-off-by: Maarten Lankhorst <maarten.lankhorst at linux.intel.com>
> >> ---
> >>  drivers/gpu/drm/i915/intel_display.c | 11 ++++++++---
> >>  1 file changed, 8 insertions(+), 3 deletions(-)
> >>
> >> diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
> >> index 1e42309ec40a..b76778d76035 100644
> >> --- a/drivers/gpu/drm/i915/intel_display.c
> >> +++ b/drivers/gpu/drm/i915/intel_display.c
> >> @@ -15421,6 +15421,7 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
> >>  		WARN_ON(drm_atomic_set_mode_for_crtc(crtc->base.state, NULL) < 0);
> >>  		crtc->base.state->active = crtc->active;
> >>  		crtc->base.enabled = crtc->active;
> >> +		crtc->base.state->connector_mask = 0;
> >>  
> >>  		/* Because we only establish the connector -> encoder ->
> >>  		 * crtc links if something is active, this means the
> >> @@ -15456,20 +15457,24 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
> >>  {
> >>  	struct intel_connector *connector;
> >>  	struct drm_device *dev = encoder->base.dev;
> >> +	struct drm_crtc *crtc = encoder->base.crtc;
> >>  	bool active = false;
> >>  
> >>  	/* We need to check both for a crtc link (meaning that the
> >>  	 * encoder is active and trying to read from a pipe) and the
> >>  	 * pipe itself being active. */
> >> -	bool has_active_crtc = encoder->base.crtc &&
> >> -		to_intel_crtc(encoder->base.crtc)->active;
> >> +	bool has_active_crtc = crtc && crtc->state->active;
> >>  
> >>  	for_each_intel_connector(dev, connector) {
> >>  		if (connector->base.encoder != &encoder->base)
> >>  			continue;
> >>  
> >>  		active = true;
> >> -		break;
> >> +		if (!has_active_crtc)
> >> +			break;
> >> +
> >> +		crtc->state->connector_mask |=
> >> +			1 << drm_connector_index(&connector->base);
> > I still think this is the wrong place. Imo this should be done in
> > intel_modeset_update_connector_atomic_state. It'd be great if we could
> > somehow share the logic with drm_atomic_set_crtc_for_connector even, but
> > that's probably over the top.
> >
> No it should be done sooner. I want to be able use it anywhere in the
> .crtc_disable calls without worrying about it..

Well I don't want to split things up all over. Atm our state recover is a
complete mess, and we need to start recovering some order in it. Updating
related things at completely different places without even a comment
stating why that's required is imo a no-go.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch